weird hard drive problems

weird hard drive problems

Post by Shashank Misr » Fri, 12 Feb 1999 04:00:00



Three thoughts-

1. Boot off an emergency floppy (one with a bootable kernel on it, e.g. boot/root
floppies from slack install) or switch to whatever runlevel on slackware is the
debugging runlevel (look in /etc/inittab to figure out which one that might be). With
/dev/hda5 NOT mounted, run 'fsck.ext2 /dev/hda5' This will fsck the partition to see
if there are any hardware problems physically with the drive.

2. Check which dma chip you have on your motherboard. there are a few (it's in the
kernel docs... somewhere... off the top of my head, I know they are in the hints tabs
if you make menuconfig the kernel) that require special bug work-arounds.

3. Check BIOS settings- my BIOS has a setting where you can reduce IDE speed (that's
just what the entry is- I haven't the slightest what it actually does). A company in
Redmond, WA makes an OS that geeks out on me if I set my bus speed to high.

I am assuming you compiled the kernel carefully- e.g. using appropriate versions of
all the relevant programs....

s

 
 
 

weird hard drive problems

Post by Juergen Fiedle » Sat, 13 Feb 1999 04:00:00


Hi,

I am running Slackware 3.6 with a hand compiled kernel and I'm running
into the following problem:

Sometimes when I try to access data on /dev/hda5, I get a bunch of
very weird errors. They look like this:

Quote:>-------SNIP-------<

hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hda: dma_intr: error=0x40 { UncorrectableError }, LBAsect=2776834, sector=1901826
end_request: I/O error, dev 03:05, sector 1901826

Quote:>-------SNIP-------<

This particular error occurred when I tried to compile a file that was
located on /dev/hdb5. I first saw it during the installation of
Slackware and thought I had bypassed it by formatting the partition
by hand. Well, obviously, I was wrong.

/dev/hda5 resides on an old 1.5GB Maxtor IDE drive. /dev/hda1 is 300MB in
size and contains the root directory. /dev/hda2 is 128MB and contains the
swap partition. /dev/hda5 takes the rest of the room (about 1.1GB).
That's all of the information I can think of at the moment.
Does anyone have any idea what went wrong?

TIA,
Juergen

 
 
 

1. WEIRD hard drive problem.

Okay, I'm throwing this one out.  I've got a Conner 1.2 gb HD that's
been running just fine in my system for a while now.  1 gig for DOS,
200 megs for Linux.  I decided to up my Linux space recently, and
sunk $60 into a Conner 420.  It's correctly jumpered for slave.  I
slapped it into my Primary IDE chain and set the BIOS on it.  
Autodetection worked fine.  I exited BIOS and booted the machine.  It
hung when it looked for the drive, then squaked with a "HD Controller
Failure"  

I hit 'F1' and resumed.  Both LINUX and Win95 found the drive.  I
fdisked it in Linux and used Win95 to format the DOS partition on it
(a fifty megger).  No problem.  Linux FDISK didn't report any
problems either.  

But when I went to use 'e2fsck' (the version shipped with Slackware
3.0) it dumped me out with a 'bad magic number' error and squabbled
something about the superblock.  Eh?

Anyway, to get around the delay at boot I took the geometry out of
the BIOS, and the drive itself has been moved; it's now on my
secondary controller, slave to my CD-ROM master.  (/dev/hdd1-hdd4)

Any thinks?  Anyone?

2. Firewall in FreeBSD 3.0

3. Weird Problem: Hangs on boot from hard drive

4. eth0: Too much work at interrupt, status0x01 or 0x40 or 0x41 ??

5. copying hard drive to hard drive

6. no route to host problem

7. ATI mouse - also weird hard drive messages

8. RedHat NOT Recognizing FD Controller

9. Can SUSE 7.2 be installed from a local hard drive to a local hard drive

10. How to merge two hard drive into one logical hard drive for Linux

11. Please Help: Weird Hard Drive Behavior!

12. very weird: entire system "blocks" during hard-drive IO

13. Weird issues cloning a Linux system to new hard drive